This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: fft.texi, cheb.texi
- To: sean at sychron dot com
- Subject: Re: fft.texi, cheb.texi
- From: Brian Gough <bjg at network-theory dot co dot uk>
- Date: Fri, 7 Sep 2001 13:49:08 +0100 (BST)
- Cc: gsl-discuss at sources dot redhat dot com
- References: <01090709000100.23700@longmorn.sychron.com>
Sean Harbour writes:
> 1. --
> I noticed that the example in the file "doc/fft.texi" contains
> references to an undefined variable "z", should this be "data"
> (first few lines).
> 2. --
> Also, in the example in file "doc/cheb.texi"
> The step size is a power of 10 and doesn't have an exact binary
> representation. "0.00125" may be better (at least give it a
> fighting chance) or similar -- people will use this as reference
> code.
Thanks, now updated.
> 3. --
> Again in "doc/cheb.texi" "order" is used where I would expect
> "degree" to be used, and "term" where I would expect "order" to be
> used. That is, the usual relationship between degree and order is
> that "order = degree+1" (as in a polynomial of degree 2 has order
> 3, a quadratic).
I've corrected one place that was clearly wrong but I'll let the
original author look at the others. I think the order refers to the
highest chebshev polynomial in the sum T0(x) == zeroth order approx,
T1(x) == first order approx, T2(x) == second order approx. Maybe that
is physicist's lingo so it didn't seem odd to me.
regards
Brian Gough